Error component model for networks

Specification and testing

Abstract

This paper develops a subgraph network random effects error components for network data regression models. In particular, it allows for edge and triangle specific components, which serve as a basal model for modeling network effects. It then evaluates the potential effects of ignoring network effects in the estimation of the variance-covariance matrix. It also proposes consistent estimator of the variance components and Lagrange Multiplier tests for evaluating the appropriate model of random components in networks. Monte Carlo simulations show that the tests have good performance in finite samples. It applies the proposed tests to the Call interbank market in Argentina.